View Single Post
Old 05-23-2019, 02:13 PM
Sam Stone is offline
Charter Member
Join Date: Jun 1999
Posts: 28,479
Originally Posted by SamuelA View Post
Gigo, how many years do you think we are from mass deployment of robotic systems able to do any repetitive task with a reliably measurable short-term goal*?

With such systems, mass deployment to 100% capacity would just be a matter of time.

*I can define this pretty exactly if you need
I'm not GigoBuster, but it sounds like you are suggesting mass robotic manufacturing of solar panels from construction to transportation to in situ mounting and hookup and all the rest. Is that correct?

If so, the approximate answer is 'never', in the sense that if we ever get to the point where we could do something like that, the world will be so different that the arguments we are having here will be moot. For one thing, such automated manufacturing would apply to everything from nuclear plants to CO2 scrubbers.

Even inside factories we haven't figured out how to automate 100% of the activities. Even auto plants which have been automated heavily still require many workers to do the stuff the machines just aren't capable of doing. I have visited a lot of automated plants, since I used to work in factory automation. All of them are teeming with humans. Remember when Elon Musk announced that he was going to be building 'lights out' automated factories with no people in them? That didn't happen. He grossly overestimated our ability to automate things.

And the factory floor is the *easiest place to automate. Today there is little to no on-site automation of installations, because there isn't enough site-to-site consistency to automate a process. Installing a solar panel on a roof always results in problems that it takes human judgement to solve, just as installing plumbing or wiring does. You don't see a lot of robots on job sites, and you won't for a long time unless they are doing menial tasks (like a robotic mixture stirrer, or a robotic nail gun or something).

The day may eventually come when you can order an installation of some sort and have robots deliver it and install it, but if that day comes it won't be for decades, and it will bring with it so many changes to the world that all the numbers we're using today are moot.

Now, you could see technologies like 3D printing bring down costs, and those are available now and could be scaled up to 3D print buildings or large concrete or steel support structures and the like. But then, those technologies could also be used to build other types of energy installations.